[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: NT restart without regular shutdown
- To: Mailing List Recipients <email@example.com>
- Subject: Re: NT restart without regular shutdown
- From: firstname.lastname@example.org (Roy Gwinn)
- Date: Thu, 21 Jan 1999 15:08 +0000 (GMT)
- CC: email@example.com
- Delivered-To: pcisig@teleport.COM
- In-Reply-To: <36A6E441.firstname.lastname@example.org>
- Old-X-Envelope-From: email@example.com
- Reply-To: firstname.lastname@example.org
- Resent-Date: Thu, 21 Jan 1999 07:39:53 -0800
- Resent-From: email@example.com
- Resent-Message-ID: <"eTHDx3.0.at2.d8qfs"@electra.znyx.com>
- Resent-Sender: firstname.lastname@example.org
> Device drivers must quieten their devices at first because devices
> can assert INTx#.
There shouldn't be any way out of a Win NT 'blue screen' other than
by pressing the hardware reset button. This will assert PCI RESET and
so disable all PCI device interrupts.
Some BIOSes fail to assert the PCI RESET line on a soft restart. So
if you can do a soft restart from the blue screen (which I doubt)
this might leave you with an active interrupting device and no
installed driver for it.
> what will happen if driver A receives the interrupt
> requested by device B?
Ordinarily this won't happen, as driver B must load before device B's
interrupts are enabled.
> in my application, driver A dies. Why?
> How can I solve this problem?
I can recommend NuMega's Soft-ICE for Windows NT for investigating
this kind of thing.